[Bug-apl] Merging of Emacs module native code

2014-03-15 Thread Elias Mårtenson
Hello Jürgen, Prior to the release of 1.3, could you merge the code for the native module from the Emacs mode? The code is pretty stable right now and ready for 1.3. Like last time, it's in the *native* branch of the Git repository: https://github.com/lokedhs/gnu-apl-mode/tree/native Regards, E

[Bug-apl] OpenMP patch, test harness and benchmark

2014-03-15 Thread David B. Lamkins
See attached files. Configured using ASSERT_LEVEL_WANTED=0, the execution ratio improved slightly to about 2:1. Note that this test uses a wider range of ravel lengths. The execution time ratio is pretty much constant, falling off only for the smallest (100 element) ravels. I did monitor the sys

Re: [Bug-apl] Performance optimisations: Results

2014-03-15 Thread Juergen Sauermann
Hi, one more thing is proper .configure (see README-2-configure). In particular VALUE_CHECK_WANTED=no and ASSERT_LEVEL_WANTED=0, otherwise you get a sequential time component proportional to the result size. /// Jürgen On 03/14/2014 07:56 PM, David Lamkins wrote: Hmm, I hadn't thought of

Re: [Bug-apl] "Largest integer" isnt' really largest?

2014-03-15 Thread Juergen Sauermann
Hi Fred, I have changed the implementation of +, -, ×, ÷, and ⋆ to use integer arithmetic as long as the result remains in the range -920 to 920 (i.e. almost full 63-bit precision), see SVN 166. Documentation update will be in SVN 167 and the info document in the